![]() ![]() This document does not contain explanations or purposes of the libraryįunctions, nor does it contain details of the approach used in the CUDA Library components that give expert programmers direct control of all Is a lower-level explicit programming model with substantial runtime Compared to the NVIDIAĪccelerator and OpenACC directives-based model and compilers, CUDA Fortran CUDA Fortran includes aįortran 2003 compiler and tool chain for programming NVIDIA GPUs usingįortran, and is an analog to NVIDIA's CUDA C compiler. OpenMP specification, refer to the OpenMP organization's website.ĬUDA Fortran is a small set of extensions to Fortran that supports and isīuilt upon the CUDA computing architecture. Host code differ only in syntax compared to those used in OpenACC.įor general information about using OpenMP and to obtain a copy of the The methods for passing device arrays to library functions from The OpenMP target offload capabilities are similar in many respects to Used to specify parallel execution from Fortran (and other languages). Programming interface (API), and a set of environment variables that can be OpenMP is a specification for a set of compiler directives, an applications See the OpenACC website for more information about the OpenACC The OpenACC API was designed and is maintainedīy an industry consortium. Loops and regions of code for offloading from a host CPU to an attachedĪccelerator, such as a GPU. The OpenACC Application Program Interface is a collection of compilerĭirectives and runtime routines that allows the programmer to specify NVTX, an API for annotating application events, code ranges, and resources.NVSHMEM, a library implementation of OpenSHMEM on GPUs.NCCL, a collective communications librarys.cuTENSOR, a library for tensor primitive operations.cuSOLVER, a library of equation solvers used with dense or other matrices.cuSPARSE, a library of linear algebra routines used with sparse matrices.cuRAND, a library for random number generation.cuFFT, a library of Fast Fourier Transform (FFT) routines.Libraries which NVIDIA provides pre-built interface modules for, and which Programming models, or from NVIDIA CUDA Fortran. It can be used from Fortran code using the OpenACC or OpenMP This document provides a reference for calling CUDA Library functions from ![]() The organization of this document is as follows: Introduction contains a general introduction to Fortran interfaces, OpenACC, CUDA Fortran, and CUDA Library functions BLAS Runtime Library APIs describes the Fortran interfaces to the various cuBLAS libraries FFT Runtime Library APIs describes the module types, definitions and Fortran interfaces to the cuFFT library Random Number Runtime APIs describes the Fortran interfaces to the host and device cuRAND libraries Sparse Matrix Runtime APIs describes the module types, definitions and Fortran interfaces to the cuSPARSE Library Matrix Solver Runtime APIs describes the module types, definitions and Fortran interfaces to the cuSOLVER Library Tensor Primitives Runtime APIs describes the module types, definitions and Fortran interfaces to the cuTENSOR Library NVIDIA Collective Communications Library APIs describes the module types, definitions and Fortran interfaces to the NCCL Library NVSHMEM Communication Library APIs describes the module types, definitions and Fortran interfaces to the NVSHMEM Library NVTX Profiling Library APIs describes the module types, definitions and Fortran interfaces to the NVTX API and Library Examples provides sample code and an explanation of each of the simple examples. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |